Quantum Computing: A Step Towards a Promising Future for Quantum Processing | Nature | Nature’s portfolio

This week’s paper reports that a quantum processor can outperform classical computation without implementing error correction.natureIt has been demonstrated that a 127-qubit IBM processor is capable of generating a highly entangled quantum state and measured the expected value of this quantum state, which is an estimated average of the results of repeated experiments. This is beyond the capabilities of the best classical calculus currently available. This proof-of-concept indicates that quantum processors may soon be useful for some specific computational operations without fault tolerance (the behavior of a quantum computer that can be quickly avoided or corrected and controlled). Meanwhile, achieving fault-tolerant computing will likely take many more years.

One of the main goals of quantum computing is to perform certain tasks more efficiently than conventional computers, and achieving this goal requires tackling a certain number of practical challenges. For example, keeping the error rate low and eliminating quantum “noise” (perturbations from the platform and environment) as quantum computers get larger. It reduces errors and noise and removes the advantage of quantum computing over classical computation. Fault tolerance is not achieved with current technologies. While current quantum processors have been shown to outperform classical processors in performing specific but intentionally designed problems, currently or in the near future troublesome quantum computers, there is an ongoing debate about whether it is sufficient, for example, to perform quantum computations that can be useful for research purposes.

Andrew Edens, Yong-Seok Kim, Abhinav Kadala and colleagues present evidence that this quantum chip can generate, manipulate and measure complex quantum states whose properties cannot be reliably estimated by classical approximations. The demonstration experiment conducted this time was at the point where quantum computers can be used to solve some specific problems (such as studying physical models) that are difficult to solve on classical computers, even without error correction. The authors performed experiments using a 127-qubit processor to run a 60-layer-deep circuit containing about 2,800 binary-qubit gates (the quantum equivalent of logic gates in classical computers). Such quantum circuits generate large, highly entangled quantum states, which classical computers lack the ability to reliably reproduce by numerical approximation. The authors show that a quantum computer can be used to accurately estimate the properties of these quantum states by measuring the predictions. The ability to generate and measure such large quantum states without introducing too many errors to interfere with the calculations is due to the high quality of the chips manufactured and the analysis that corrects for noise, due to the presence of a post-processing method.

In a companion article for News & Views, Göran Wendin and Jonas Bylander conclude, “The primary quantum advantage in this work lies in size, not speed. With 127 qubits, the memory of a classical computer is not enough.” The problem is in the area of ​​​​a huge country that does not exist.

Two: 10.1038 / s41586-023-06096-3

Back to the “Featured Papers” list